Con Kolivas
a4ec961ecc
We can queue all the necessary work without hitting frequent stales now with the time and string stale protection active all the time.
...
This prevents a pool being falsely labelled as not providing work fast enough.
13 years ago
Con Kolivas
81aedc972a
Add message about needing one server.
13 years ago
Con Kolivas
f2f0ba8024
Revert "Revert "Since we roll work all the time now, we end up staging a lot of work without queueing, so don't queue if we've already got staged work.""
...
This reverts commit cea1cf6cc0
.
This wasn't eht cause of the false positives.
13 years ago
Con Kolivas
cea1cf6cc0
Revert "Since we roll work all the time now, we end up staging a lot of work without queueing, so don't queue if we've already got staged work."
...
This reverts commit b81077f36a
.
This creates too many false positives for pool not providing work fast enough.
13 years ago
Con Kolivas
5a2cf5a6b1
Get start times just before mining begins to not have very slow rise in average.
13 years ago
Con Kolivas
b643b56a95
Allow LP to reset block detect and block detect lp flags to know who really came first.
13 years ago
Con Kolivas
48f07d9219
Update to latest sse2 code from cpuminer-ng.
13 years ago
Con Kolivas
73c98e1e79
Check if there is more than one work item queued before complaining about a slow pool.
13 years ago
Con Kolivas
dbf0a1366d
Use the new hashes directly for counts instead of the fragile counters currently in use.
13 years ago
Con Kolivas
0899ee86ae
Only consider pool slow to respond if we can't even roll work.
13 years ago
Con Kolivas
7407e887f6
Update configure.ac for newer autoconf tools.
13 years ago
Con Kolivas
6197ff2009
Remove silly debugging output.
13 years ago
Con Kolivas
9ab15a8266
Merge branch 'hash'
13 years ago
Con Kolivas
93f4163aca
Create a hash list of all the blocks created and search them to detect when a new block has definitely appeared, using that information to detect stale work and discard it.
13 years ago
Con Kolivas
b81077f36a
Since we roll work all the time now, we end up staging a lot of work without queueing, so don't queue if we've already got staged work.
13 years ago
Con Kolivas
bf3033e0f1
Make restarting of GPUs optional for systems that hang on any attempt to restart them.
...
Fix DEAD status by comparing it to last live time rather than last attempted restart time since that happens every minute.
13 years ago
Con Kolivas
666fcc3f55
Move staged threads to hashes so we can sort them by time.
13 years ago
Con Kolivas
bca84c3ba0
Make restarting of GPUs optional for systems that hang on any attempt to restart them.
...
Fix DEAD status by comparing it to last live time rather than last attempted restart time since that happens every minute.
13 years ago
Con Kolivas
2630c6d68d
Bump version to 1.5.8.
13 years ago
Con Kolivas
d9accc4846
Put a lower limit on the nonce increment in cpu mining.
13 years ago
Con Kolivas
175ef7d936
Update news.
13 years ago
Con Kolivas
f6591379fb
Minimise how much more work can be given in cpu mining threads each interval.
13 years ago
Con Kolivas
5acdbec47d
Make the fail-pause progressively longer each time it fails until the network recovers.
13 years ago
Con Kolivas
b4d6b11b9a
Only display the lagging message if we've requested the work earlier.
13 years ago
Con Kolivas
ced4a9ac92
Clean up the pool switching to not be dependent on whether the work can roll or not by setting a lagging flag and then the idle flag.
13 years ago
Con Kolivas
088ee2fa29
Only use one thread to determine if a GPU is sick or well, and make sure to reset the sick restart attempt time.
13 years ago
Con Kolivas
02e126f42d
The worksize was unintentionally changed back to 4k by mistake, this caused a slowdown.
13 years ago
Con Kolivas
cc9d5f5927
Bump version number to 1.5.7.
13 years ago
Con Kolivas
f465397fa6
Update news for 1.5.7.
13 years ago
Con Kolivas
6d10ef2f6e
Bump version numbers of kernels to indicate slightly different versions.
13 years ago
Con Kolivas
dfec430cb8
Make the tcp setsockopts unique to linux in the hope it allows freebsd et. al to compile.
13 years ago
Con Kolivas
7c6dfaccf1
Display which longpoll failed and don't free the ram for lp_url since it belongs to the pool hdr path.
13 years ago
Con Kolivas
6f741597f8
Use any longpoll available, and don't disable it if switching to a server that doesn't have it.
13 years ago
Con Kolivas
b358bdc07e
Never automatically disable any pools but just specify them as idle if they're unresponsive at startup.
13 years ago
Con Kolivas
324d849bf2
Attempt to restart a GPU once every minute while it's sick.
...
Don't kill off the reinit thread if it fails to init a GPU but returns safely.
Only declare a GPU dead if there's been no sign of activity from the reinit thread for 10 mins.
13 years ago
Con Kolivas
1b228ee3d4
Re-enable devices immediately after ther old threads would have been killed off in case the first attempt to restart them fails but we still want them emabled.
13 years ago
Con Kolivas
75018d0d48
Release cl resources should the gpu mining thread abort.
13 years ago
Con Kolivas
f062765c94
Explicitly link in ws2_32 on the windows build and update README file on how to compile successfully on windows.
13 years ago
Con Kolivas
bd79a61c43
Move poclbm to new branch optimisation as well.
13 years ago
Con Kolivas
aaa2e19e0e
Add one more instruction to avoid one branch point in the common path in the cl return code.
...
Although this adds more ALUs overall and more branch points, the common path code has the same number of ALUs and one less jmp, jmps being more expensive.
13 years ago
Con Kolivas
a754cc3f0c
Suppress correct log output when display per-device status.
13 years ago
Con Kolivas
4beade3772
Retain the program immediately after it's created from source.
13 years ago
Con Kolivas
082e20df5f
Explicitly tell the compiler to retain the program to minimise the chance of the zero sized binary errors.
13 years ago
Con Kolivas
6f769e0f8f
Add the ability to enable/disable per-device stats on the fly and match logging on/off.
13 years ago
Con Kolivas
e414490bf7
Don't display rolling status if per device stats is enabled.
13 years ago
Con Kolivas
2798b8c549
Merge branch 'master' of github.com:ckolivas/cgminer
13 years ago
Con Kolivas
d89a6c57b1
Since we check roll time per work item now, it need only be debug log level.
13 years ago
Con Kolivas
64e7cdd61b
The new phatk kernel needs a different nonce passed according to how many vectors are in use.
...
This fixes breakage that otherwise occurs when 1 or 4 vectors are chosen.
13 years ago
Con Kolivas
3926ad96c8
Merge pull request #36 from znort987/per-device-stats
...
Add per-device statics log output
13 years ago
Znort 987
e49bfbf2ec
Add per-device statics log output
13 years ago